6. kubeadm-install-v1.8.3

https://kubernetes.io/docs/setup/independent/create-cluster-kubeadm/

6.1. env

192.168.31.120 km master
192.168.31.119 kn1 node
192.168.31.118 kn2 node

6.2. Initializing your master

kubeadm init --pod-network-cidr=10.244.0.0/16

如果遇到类似下面错误

[preflight] Some fatal errors occurred: Port 10250 is in use /etc/kubernetes/manifests is not empty /var/lib/kubelet is not empty

则，参考 https://github.com/kubernetes/kubernetes/issues/37063 运行下面命令：

kubeadm reset
systemctl start kubelet.service

之后，再次运行

kubeadm init --pod-network-cidr=10.244.0.0/16

被墙了，出不去，我了个去，怎么办？

https://mritd.me/2016/10/29/set-up-kubernetes-cluster-by-kubeadm/#21安装包从哪来

好吧，那就去 hub.docker.com 中配置吧

6.3. 找到所有要配置的 image

找 /etc/kubernetes/manifests/

root@km:~# cd /etc/kubernetes/manifests/
root@km:/etc/kubernetes/manifests# ls
etcd.yaml  kube-apiserver.yaml  kube-controller-manager.yaml  kube-scheduler.yaml
root@km:/etc/kubernetes/manifests#
root@km:/etc/kubernetes/manifests# cat *.yaml | grep image
    image: gcr.io/google_containers/etcd-amd64:3.0.17
    image: gcr.io/google_containers/kube-apiserver-amd64:v1.8.3
    image: gcr.io/google_containers/kube-controller-manager-amd64:v1.8.3
    image: gcr.io/google_containers/kube-scheduler-amd64:v1.8.3
root@km:/etc/kubernetes/manifests#

找源码

https://github.com/kubernetes/kubernetes/tree/master/cmd/kubeadm

root@km:~/kubernetes.20171116/cmd/kubeadm# git clone https://github.com/kubernetes/kubernetes kubernetes.20171116
root@km:~# cd kubernetes.20171116
root@km:~/kubernetes.20171116# cd cmd/kubeadm/
root@km:~/kubernetes.20171116/cmd/kubeadm# grep gcr.io -r ./
./app/apis/kubeadm/v1alpha1/defaults.go:    DefaultImageRepository = "gcr.io/google_containers"
./app/phases/selfhosting/selfhosting_test.go:    image: gcr.io/google_containers/kube-apiserver-amd64:v1.7.4
./app/phases/selfhosting/selfhosting_test.go:        image: gcr.io/google_containers/kube-apiserver-amd64:v1.7.4
./app/phases/selfhosting/selfhosting_test.go:    image: gcr.io/google_containers/kube-controller-manager-amd64:v1.7.4
./app/phases/selfhosting/selfhosting_test.go:        image: gcr.io/google_containers/kube-controller-manager-amd64:v1.7.4
./app/phases/selfhosting/selfhosting_test.go:    image: gcr.io/google_containers/kube-scheduler-amd64:v1.7.4
./app/phases/selfhosting/selfhosting_test.go:        image: gcr.io/google_containers/kube-scheduler-amd64:v1.7.4
./app/phases/selfhosting/selfhosting_test.go:    - image: gcr.io/google_containers/busybox
./app/phases/selfhosting/selfhosting_test.go:        "image": "gcr.io/google_containers/busybox"
./app/phases/selfhosting/selfhosting_test.go:  - image: gcr.io/google_containers/busybox
./app/phases/upgrade/staticpods_test.go:imageRepository: gcr.io/google_containers
./app/util/template_test.go:        validTmplOut = "image: gcr.io/google_containers/pause-amd64:3.0"
./app/util/template_test.go:        doNothing    = "image: gcr.io/google_containers/pause-amd64:3.0"
./app/util/template_test.go:                                ImageRepository: "gcr.io/google_containers",
./app/util/template_test.go:                                ImageRepository: "gcr.io/google_containers",
./app/images/images_test.go:        gcrPrefix   = "gcr.io/google_containers"
./app/constants/constants.go:       DefaultCIImageRepository = "gcr.io/kubernetes-ci-images"
root@km:~/kubernetes.20171116/cmd/kubeadm#

最后，居然什么也没有找到。哎呀。好吧。

回到 https://gitee.com/tomt/tom_docker_registry_push_pull.git ，看了一下，也就是这4个需要更新一下。

gcr.io/google_containers/kube-apiserver-amd64:v1.8.3
gcr.io/google_containers/kube-proxy-amd64:v1.8.3
gcr.io/google_containers/kube-scheduler-amd64:v1.8.3
gcr.io/google_containers/kube-controller-manager-amd64:v1.8.3

好了。明确了image, 那就做吧。

github hub.docker.io 配置 image

注意，在 hub.docker.io 下，自动构建仓库，时，取 github的仓库，对这个仓库名称有限制

  • 不能大于3个 “-” 号
  • 长度不能太长

这样呢，gcr.io/google_containers/kube-controller-manager-amd64:v1.8.3， 在github上本来要创建名称为 dockerlibraryk8s-kube-controller-manager-amd64 ，那这个不符合要求的。 怎么办？改个名称咯。 我把它修改成 dockerlibraryk8s-kube-cm-amd64了。 记得最后，pull 下来的时候，docker tag 一下。

root@km:~/tom_k8s_kubernetes_install/k8/docker-library-k8s-v1.8.3# ls
dockerlibraryk8s.sh  k8s-v1.8.3.txt
root@km:~/tom_k8s_kubernetes_install/k8/docker-library-k8s-v1.8.3# cat dockerlibraryk8s.sh
#images=(heapster-influxdb-amd64:v1.3.3 heapster-amd64:v1.4.0 heapster-grafana-amd64:v4.4.3)
#images=(heapster-grafana-amd64:v4.4.3)
for imageName in `cat k8s-v1.8.3.txt`; do
echo $imageName
echo ${imageName%:*}
echo ${imageName#*:}
imageNamelast=dockerlibraryk8s-${imageName%:*}
echo docker pull tomtsang/$imageNamelast
docker pull tomtsang/$imageNamelast
echo docker tag tomtsang/$imageNamelast gcr.io/google_containers/$imageName
docker tag tomtsang/$imageNamelast gcr.io/google_containers/$imageName
echo docker rmi tomtsang/$imageNamelast
docker rmi tomtsang/$imageNamelast
done
echo "game over"

root@km:~/tom_k8s_kubernetes_install/k8/docker-library-k8s-v1.8.3# docker pull tomtsang/dockerlibraryk8s-kube-cm-amd64
Using default tag: latest
latest: Pulling from tomtsang/dockerlibraryk8s-kube-cm-amd64
0ffadd58f2a6: Already exists
18c5c31a1ebe: Pull complete
Digest: sha256:4e738f80a607772c205ca597c1d5874ee50ac40f0a5e88ab85084fd45b684ac0
Status: Downloaded newer image for tomtsang/dockerlibraryk8s-kube-cm-amd64:latest
root@km:~/tom_k8s_kubernetes_install/k8/docker-library-k8s-v1.8.3# docker tag tomtsang/dockerlibraryk8s-kube-cm-amd64 gcr.io/google_containers/dockerlibraryk8s-kube-cm-amd64:v1.8.3
root@km:~/tom_k8s_kubernetes_install/k8/docker-library-k8s-v1.8.3# docker tag gcr.io/google_containers/dockerlibraryk8s-kube-cm-amd64:v1.8.3 gcr.io/google_containers/kube-controller-manager-amd64:v1.8.3
root@km:~/tom_k8s_kubernetes_install/k8/docker-library-k8s-v1.8.3# docker rmi tomtsang/dockerlibraryk8s-kube-cm-amd64
Untagged: tomtsang/dockerlibraryk8s-kube-cm-amd64:latest
Untagged: tomtsang/dockerlibraryk8s-kube-cm-amd64@sha256:4e738f80a607772c205ca597c1d5874ee50ac40f0a5e88ab85084fd45b684ac0
root@km:~/tom_k8s_kubernetes_install/k8/docker-library-k8s-v1.8.3#

使用 http://git.oschina.net/tomt/tom_k8s_kubernetes_install 这个仓库吧。

root@km:~/tom_k8s_kubernetes_install/k8/docker-library-k8s-v1.8.3# git remote -v
origin      http://git.oschina.net/tomt/tom_k8s_kubernetes_install (fetch)
origin      http://git.oschina.net/tomt/tom_k8s_kubernetes_install (push)
root@km:~/tom_k8s_kubernetes_install/k8/docker-library-k8s-v1.8.3# ls
dockerlibraryk8s.sh  k8s-v1.8.3.txt
root@km:~/tom_k8s_kubernetes_install/k8/docker-library-k8s-v1.8.3# pwd
/root/tom_k8s_kubernetes_install/k8/docker-library-k8s-v1.8.3
root@km:~/tom_k8s_kubernetes_install/k8/docker-library-k8s-v1.8.3# ls
dockerlibraryk8s.sh  k8s-v1.8.3.txt
root@km:~/tom_k8s_kubernetes_install/k8/docker-library-k8s-v1.8.3# ./dockerlibraryk8s.sh

到现在为止，应该 4 个 image都成功pull 到了 master节点了。

6.4. docker registry push

因为我们后面要在 node 节点上使用，所以干脆就直接 push 到 docker registry 去吧。

6.5. kubeadm-init-use-local-image

参看 kubeadm-init-use-local-image.rst 文件。

好像没成功

6.6. kubeadm init

root@km:~# export
...
declare -x http_proxy="http://192.168.31.239:8118/"
declare -x https_proxy="http://192.168.31.239:8118/"
declare -x no_proxy="localhost,127.0.0.1,192.168.31.120,10.96.0.10,github.com,ubuntu.com"

root@km:~# kubeadm init --pod-network-cidr=10.244.0.0/16 --skip-preflight-checks
....
....
....
To start using your cluster, you need to run (as a regular user):

mkdir -p $HOME/.kube
sudo cp -i /etc/kubernetes/admin.conf $HOME/.kube/config
sudo chown $(id -u):$(id -g) $HOME/.kube/config

You should now deploy a pod network to the cluster.
Run "kubectl apply -f [podnetwork].yaml" with one of the options listed at:
http://kubernetes.io/docs/admin/addons/

You can now join any number of machines by running the following on each node
as root:

kubeadm join --token ce4253.8322cc2590378260 192.168.31.120:6443 --discovery-token-ca-cert-hash sha256:bb0b9ef27e5ffef06776ca10a87ed548cefedc703ddaf904316c87d4a7f3655d

有image, 而不能启动docker container.